siddk/entity-network

Tensorflow implementation of "Tracking the World State with Recurrent Entity Networks" [https://arxiv.org/abs/1612.03969] by Henaff, Weston, Szlam, Bordes, and LeCun.

34
/ 100
Emerging

This project helps AI researchers and practitioners efficiently track dynamic information by maintaining a 'world-state' based on incoming data. It takes sequences of information (like sentences from a story) as input and outputs updated memory cells that represent different concepts or entities. The primary users are those working on advanced natural language processing and question-answering systems.

No commits in the last 6 months.

Use this if you need a robust way to update a system's understanding of various concepts as new information arrives, especially in complex environments like conversational AI or knowledge tracking.

Not ideal if your task involves simple classification or regression problems where maintaining a dynamic, interpretable 'world-state' isn't a core requirement.

natural-language-understanding question-answering conversational-ai knowledge-representation dynamic-memory-systems
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 8 / 25
Community 18 / 25

How are scores calculated?

Stars

57

Forks

16

Language

Python

License

Last pushed

Mar 08, 2017

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/siddk/entity-network"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.